NEW!: Postdoc Position Opening

Current Projects

• FreeLoader: Distributed Storage Scavenging

  • Leading an effort to build terabytes of cache/scratch space for transient, bulk, immutable data from loosely connected, commodity, user desktop storage resources.
  • Co-PI on a broader effort to provide reliability, availability and serviceability for terascale supercomputers. I am delving into the data availability aspect (10/2004 - 10/2006).

• Dynamic Staging Architecture for Accelerating I/O Pipelines

  • Leading an effort to build a staging area using SSDs on large supercomputers that can be used to accelerate I/O and post-processing pipelines of large (e.g., 100,000-core) jobs.

• CenterSS: HPC Center Storage as a System

  • An Integrated Approach to Machine-Room Storage Management
  • Research topics addressed: timely staging and offloading, coordination with computation, scratch as a cache, transient data recovery, etc.
  • PI on an effort to build robust storage management for the supercomputer machine-room and beyond (LDRD 10/2006 - 10/2008).

• stdchk: A Checkpoint Storage System for HPC Applications

• Leading the development of a checkpoint storage system for HPC apps at all granularity (loosely coupled to tightly coupled, i.e., from a desktop grid to clusters to large supercomputers)

Selected Recent Publications

• H. Monti, A.R. Butt, S.S. Vazhkudai, "/Scratch as a Cache: Rethinking HPC Center Scratc h Storage", Proceedings of the 23rd Int'l Conference on Supercomputing 2009 (ICS'09), Yorktown Heights, New York, June 2009. pdf
  
  
• H. Monti, A.R. Butt, S.S. Vazhkudai, "Timely Offloading of Result-Data in HPC Centers", Proceedings of the 22nd ACM Int'l Conference on Supercomputing (ICS'08), Kos, Greece, June 2008. pdf
  
  
• S.A. Kiswany, M. Ripeanu, S. S. Vazhkudai, A. Gharaibeh, "stdchk: A Checkpoint Storage System for Desktop Grid Computing", Proceedings of the 28th Int'l Conference on Distributed Computing Systems (ICDCS 2008), Beijing, China, June 2008. pdf
  
  
• Z. Zhang, C. Wang, S. Vazhkudai, X, Ma, G. Pike, F. Mueller, J.W. Cobb, "Optimizing Center Performance through Coordinated Data Staging, Scheduling and Recovery", Proceedings of Supercomputing 2007 (SC07): Int'l Conference on High Performance Computing, Networking, Storage and Analysis, Reno, Nevada, November 2007. pdf
  
  

Summary of Ongoing Storage Research

Top